Adhesion and Debonding Kinetics in PV Devices and Modules

Chris Bruner, Stéphanie Dupont, Ryan Brock (PV active layers) Fernando Novoa, Warren Cui (encapsulation and optics)

Scott Isaacson, Tissa Mirfakhrai (ultra barrier films)

Reinhold H. Dauskardt (dauskardt@stanford.edu)

Adhesion and Debonding Kinetics in PV Devices and Modules

Collaborators: Stephen Eglash / Stanford

David Miller and Sarah Kurtz / NREL

Mike Wiemer, Taner Bilir and Homan Yuen / Solar Junction

Peter Hebert and Jim Ermer / Spectrolab (Boeing)

Support: DoE support through the Bay Area PV Consortium, Predicts CPV

program through SunShot, and the CAMP-KAUST center.

Degradation and

Substrate

Reliability of PV Devices and Modules

Thermal cycling, stress, moisture, chemically active environmental species, and solar UV.

Uncertain degradation kinetics and reliability models.

Severe operating environments:H2O, O2, H2

other active chemicalspecies

photochemical reactions

cracking and debonding

UV Exposure

defect evolution innanomaterial

layers

surface weathering

Substrate

H2O, O2, H2 other active chemical

species

photochemical reactions

cracking and debonding

UV Exposure

defect evolution in nanomaterial

layers

surface weathering

defects in semi-conductor layers

electro-luminescent

peeling metallized grids after thermal

cycling

Device Reliability and Evolution of Defects damage propagates if mechanical stresses are large enough so that

mechanical cohesion or “driving force” G Gc J / m2 adhesion

presence of chemical species and photons, damage propagates even if environment and

G G J / m2 stress accelerates c defect evolution

Role of coupled “stress” parameters:

• mechanical stress

• temperature

• environmental species

• photons (photochemical reactions)

Substrate

H2O

hOrganic layer

Limitations of Thin-Film Adhesion Tests Indenter

Plastic zone • Indentation/Scratch Test - complex stress and deformation fields - principally qualitative results - (nano) scratch test even less quantitative

Interface Crack Substrate

Film

M P

• Peel/m-ELT Test - difficult to apply loads - plastic deformation of film - temperature complications in m-ELT

• Blister Test - compliant loading system - environmental effects - etching/machining of cavity difficult

Interface Crack

Substrate

Film

PSubstrate Film

Cavity in Substrate

Major limitations: need detailed film properties, film stress relaxation and film plasticity

principally qualitative results for all above methods!

Coh

esio

n E

nerg

y, G

(J/m

2 )

Quantitative Adhesion/Cohesion and Debond Kinetics

(temp/environment /UV effects) Adhesion/Cohesion

threshold crucial for reliability

ITO (150 nm ) PEDOT:PSS (50-100 nm)

Al (100 nm)Ca (7 nm)

P3HT/PCBM (~150 nm)

Degradation Kinetics

2

1.5

1

0.5

0 Solar Cells

Outline

• Inherent Solar Cell Thermo-Mechanical Reliability • polymer / PCBM BHJ type devices • adhesion and cohesion properties in flexible systems

• Encapsulant and Ultra-barriers in Solar Modules • synergistic effects of multiple “stressing parameters” • kinetics and lifetimes

Factors Effecting Cohesion of BHJ Layersplastic zone

• Heterojunction layer thickness – is cohesion in organic layers sensitive

to layer thickness?

• Composition of the heterojunction layer – limited bonding to fullerene – polymer/PCBM ratio makes stronger layer

P3HT PCBM Indene C60

• Molecular intercalation – manipulating the types of intermolecular

interactions

• Annealing Standard 150°C 0.5h 150°C 2h

– morphology of the BHJ layer changes with annealing

organic BHJ elastic layer

elastic substrate

 

 

   

100kDa 79kDa 53kDa 28kDa

5

20

BHJ Layer cohesion w/ P3HT MW

Low MWPlastic Zone

Debond propagation direction

GC = G0 + Gplasticity

High MW

G0

100 kDa P3HT

79kDa

53kDa 28kDa

0 50 100 150 200 250

BHJ Layer Thickness (nm)

Increasing P3HT MW 100kDa

Gplasticity

Coh

esio

n E

nerg

y, G

(J/m

2 ) c

15

10

BHJ layer

0

28 kDa P3HT

MW increases cohesion energy significantly • increased inter-chain entanglement/bridging • larger polymer domains as polymers fold

Rising slope with thickness due to plasticity Rq ~ 47 nm Rq ~ 8 nm

--

-

-

Effi

cien

cy,

(%)

Efficiency vs. Cohesion P3HT

3.2 Thin BHJ layer - low efficiency due

2.8 to lower photon harvesting

100kDa 79kDa 53kDa 28kDa

Increasing BHJ layer thickness

0 2 4 6 8 10 12 14 16 18

2.4

2.0

1.6

1.2

0.8

0.4

Thicker BHJ layer - greater degree of charge carrier recombination

+

+

+

+

a) b)

c)

Pho

ton

colle

ctio

nCohesion Energy, G (J/m2)c

ைܬௌܨܨ xߝ ൌ Distance from surface

ZnO

Synergistic Effects on Decohesion Kinetics

adhesive failure

PEDOT:PSS (30nm) Ag

P3HT:PCBM ZnO

Inert Environment G≥Gc

Effect of Moisture G<Gcc

Effect of In-Situ UVB G<Gcc

UV H2O

cohesive in PEDOT:PSS cohesive in P3HT:PCBM adhesive at the ZnO interface

Ag Ag

P3HT:PCBM

ZnO

H2O Pre-crack

PEDOT:PSS

P3HT:PCBM

ZnO

Pre-crack PEDOT:PSS

Decohesion of PEDO:PSS Transparent Electrodes D

ebon

d G

row

th R

ate,

da/

dt (m

/s) 10-3

10-4

10-5

10-6

10-7

10-8

10-9

10-10

0.0 0.4 0.8 1.2 1.6 2.0 2.410-11

Fit 1: Without Swelling Fit 2: With Swelling

80 % R.H. 60 % R.H. 40 % R.H. 5 % R.H.

60%

80%

40% 5%

Gtip = Gth = 2γ

%RH ↑ P3HT:PCBM

H2O

pre-crack

PEDOT:PSS

Glass

Ag

ZnO ITO

Moisture-assisted debond region

Pre-crack region

~ 18 nm

Debond Driving Force, G (J/m2)

80 % R.H. 60 % R.H. 40 % R.H. 5 % R.H.

Decohesion of PEDO:PSS Transparent Electrodes S.R. Dupont, F. Novoa, E. Voroshazi, R.H. Dauskardt, Adv. Func. Mat., 2013

Deb

ond

Gro

wth

Rat

e, d

a/dt

(m/s

) 10-3

10-4

10-5

10-6

10-7

10-8

10-9

10-10

0.4 0.8 1.2 1.6 2.0 2.4

Fit 1: Without Swelling Fit 2: With Swelling

Gtip = Gth = 2γ

%RH ↑

2sinh tip

o

Gda vdt

Atomistic kinetic model

60%

80%

40% 5%

10-11

0.0

Debond Driving Force, G (J/m2)

Outline

• Inherent Solar Cell Thermo-Mechanical Reliability • polymer / PCBM BHJ type devices • adhesion and cohesion properties in flexible systems

• Encapsulants and Ultra-Barriers in Solar Modules • synergistic effects of multiple “stressing parameters” • kinetics and lifetimes

Assessing UV and Environment on Debonding Kinetics

Glass Substrate

ITO

Glass Substrate

Simulated UV Exposure

polysiloxane barrier

ITO

automated growth analysis

sensitivity to < 10-11 m/s

DTS Delaminator v8.2 Lo

ad, P

Cra

ck L

engt

h, a

dP/dt

da/dt

test data Debonding Kinetics Po

explore role of:

• UV flux

• humidity, O2, OH, …

• temperature

• mechanical loading Time (s)

UV Effects on Molecular Bond Rupture UV Exposure (3.4 eV)

Cra

ck G

row

th R

ate,

da/

dt (m

/s)

10-10

10-11

10-4

10-5

10-6

10-7

10-8

10-9

No UV 0 mW/cm2

Ghv ~ 1.3 J/m2

Ghv ~ 0.8 J/m2

UV intensity 1.2 mW/cm2

0.6 mW/cm2

NkT

INCG aKdt da Ehvn

HT 2 2sinh20

0 1 2 3 4 5 6 7 Strain Energy Release Rate, G (J/m2)

EVA TPE- Bridging TPE- No Bridging0

5

10

15

20

25

30

35

40

45

50

55

Stra

in E

nerg

y R

elea

se R

ate,

G (J

/m2 )

Location of Failure

I

II

II

adhesion

Glass Substrate

EVA encapsulant

Tedlar (Polivinylfluoride)PET

EVA seed

Delamination of EVA-TPE Lamination

Parreta Antonio, et al., Solar Energy Materials & Solar Cells, 2005

TPE

EVA

• EVA Tg ~ -15°C • “ductile-to-brittle” transition

at lower T

• poor adhesive/cohesive properties at lower T

New Portable Full Panel Adhesion Back Side of Full Panel

P a

0 1000 2000 300 0

20

40

60

80

100

120

140

Load

(N) Square Cantilever Beam Adhesion

Adhesion energy, Gc, depends on:

P (delamination force) E (young modulus of the square)

h (thickness of the square)

Delaminator (v8.2) Adhesion Test System

DTS system and support: dauskardt@stanford.edu Distance (um)

Temperature Dependent EVA Debond Energy

3.0 TPE

2.5

EVA2.0

1.5

1.0

Glass Substrate

EVA encapsulant

Tedlar (Polivinylfluoride)PET

EVA seed

0.5

0.0 20 40 60 80

Temperature, T (°C)

Bridging and cavitation process

Deb

ond

Ene

rgy,

G (k

J/m

2 ) c

Tg

     

 

          

   

Adh

esiv

e E

nerg

y, G

(J/m

2 ) Encapsulant Adhesion is Highest at the Electrodes

140

120

100

80

60

40

20

0

Only Electrode (130 J/m2)

Typical Cell (14 J/m2)

No Electrode (2 J/m2)

0.0 0.2 0.4 0.6 0.8 1.0

Backsheet Encapsulant

Solar Cell + Electrodes

Encapsulant Tempered Glass

Reducing electrode surface decreases adhesive energy and affects module reliability!

Electrode Surface Ratio

20 300

Ageing Temperature Effect on Debond Energy

200

400

600

800

1000

1200

Deb

ond

Ene

rgy,

Gc (J

/m 2 )

Ageing time =1000 hrs

Unaged 40 50 60 70 80 90

Ageing Temperature (°C)

 

  

 

Backsheet Adhesion of Deployed Modules D

ebon

d E

nerg

y, G

c (J

/m2 )

1000

100

10

1 New Module

Gc= 2.8 J/m2

Gc=1.3 KJ/m2

Backsheet

EVA Seed

Solar Cell EVA encapsulant

Glass

PET Tedlar (Polivinylfluoride)

EVA encapsulant

10 years old

Temperature Effect on Debond Kinetics

0 50 100 150 200 250 300 35010-8

10-7

10-6

10-5

Deb

ond

Gro

wth

Rat

e, d

a/dt

(m/s

)

40°C

10°C20°C30°C

RH=40%

Glass Substrate

EVA encapsulant

Tedlar (Polivinylfluoride)PET

EVA seed TPE

EVA

Debond-Driving Force, G(J/m2)

Summary

• Inherent Solar Cell Thermo-Mechanical Reliability • polymer / PCBM BHJ type devices • adhesion and cohesion properties in flexible systems

• Encapsulants and Ultra-Barriers in Solar Modules • synergistic effects of multiple “stressing parameters” • kinetics and lifetimes